Siqi Fan
Logo Researcher @ AIR, Tsinghua University

I'm Siqi Fan (范嗣祺), a researcher at Institute for AI Industry Research, Tsinghua University (AIR, THU). Previously, I obtained my M.S. degree from the Institute of Automation, Chinese Academy of Sciences (CASIA) in 2022 and my B.E. degree from Shanghai Jiao Tong University (SJTU) in 2019.

My research interests lie broadly in Representation Learning in Complex Systems, spanning from the macro physical world to the micro biological world. With the goal of increasing human "Available Time" through AI technology, I primarily concentrate on

  • Autonomous Driving: boost the proportion of time available for meaningful activities
  • Biomedical Discovery: extend the total lifespan
Fueled by a passion for innovation, I am dedicated to pushing the frontiers of technology and developing impactful products.

I love music and visual arts.


Education
  • Shanghai Jiao Tong University (SJTU)
    Shanghai Jiao Tong University (SJTU)
    School of Electronic Information and Electrical Engineering
    B.E. in Automation
    Sep. 2015 - Jul. 2019
  • University of Chinese Academy of Sciences (UCAS)
    University of Chinese Academy of Sciences (UCAS)
    Institute of Automation
    M.S. in Automation
    Sep. 2019 - Jul. 2022
Research Experience
  • Autonomous System Group, Intel Labs China (ILC)
    Autonomous System Group, Intel Labs China (ILC)
    Research Intern
    Aug. 2020 - Dec. 2021
  • Institute of Automation, Chinese Academy of Sciences (CASIA)
    Institute of Automation, Chinese Academy of Sciences (CASIA)
    Student Researcher
    Sep. 2019 - Jul. 2022
  • Institute for AI Industry Research, Tsinghua University (AIR, THU)
    Institute for AI Industry Research, Tsinghua University (AIR, THU)
    Researcher
    Jul. 2022 - Present
Academic Service
Honors & Awards
  • National Scholarship
    2021
  • Pan Deng First-class Scholarship, CAS
    2022
  • Excellent Scholarship, SJTU
    2018
  • China Industrial Intelligence Challenge, State-level Outstanding Award, CAA
    2018
News
2025
Release the OpenBioMed toolkit and an agent platform for biomedicine and life science built on that. News
Mar 07
2024
Our workshop proposal Multi-Agent Embodied Intelligent Systems Meet Generative-AI Era: Opportunities, Challenges and Futures is accepted as a full day workshop @ CVPR'25. Call for Papers
Dec 21
The AI-agent system project PharmAID is launched @ FUSON PHARMA. News
Oct 23
Serve as Area Chair for 1st Workshop on Cooperative Intelligence for Embodied AI @ ECCV'24 News
Oct 12
2023
Release the 1st real-world large-scale dataset for roadside cooperative perception RCooper News
Dec 25
Our ChatDD-FM-100B ranks 1st in all four medical disciplines in C-Eval Benchmark and is the only model with an average score of more than 90. News
Sep 21
Release the 1st commercial-friendly multimodal biomedical foundation model BioMedGPT-10B. News
Aug 18
2022
Give a talk on Traffic Scenes Understanding and Simulation Testing @ ITSC'22.
Sep 18
Research Roadmap (view publications )
Autonomous Driving (AD)
  • Onboard System (Intelligent Vehicle)

    My exploration of vehicle-side environment perception began with drivable area detection (ITSC’20). Since then, I have proposed a series of perception algorithms, including an RGB 2D object detection approach designed for complex traffic environments (FII-CenterNet, T-VT’21), a semi-supervised learning approach for RGB 2D segmentation (CPCL, T-IP’22), an RGB-T segmentation approach tailored for challenging lighting conditions (SpiderMesh, TechReport’23), and a 3D segmentation approach for large-scale point clouds (SCF-Net, CVPR’21).

  • Roadside System (Intelligent Infrastructure)

    Compared to the well-studied vehicle-side perception, roadside perception faces several unique challenges, and the calibration noise caused by inevitable natural factors is one of them. Addressing that, a calibration-free BEV representation network is proposed to alleviate inaccurate calibration parameter problem (CBR, IROS’23). The development of roaside perception has been hindered by a lack of available data. On the one hand, a semantic-geometry decoupled contrastive learning framework is introduced to enhance roadside perception performance by leveraging vehicle-side data (IROAM, ICRA’25); On the other hand, the first large-scale real-world dataset for roadside cooperative perception is released, complete with benchmarks, to stimulate research in practical I2I perception (RCooper, CVPR’24).

  • Cooperative Autonomous Driving System (V2X)

    Cooperative perception can significantly enhance individual perception performance by providing additional viewpoints and expanding the sensing field. A scene-level feature cooperative perception approach is proposed (EMIFF, ICRA’24). To enable interpretable, instance-level, and flexible feature interactions, the concept of query cooperation is introduced, along with a cooperative perception framework that allows query streams to flow among agents (QUEST, ICRA’24). Additionally, motion forecasting can also benefit from learning cooperative trajectory representations (NeurIPS’24). Beyond focusing on improving individual modules, a pioneering end-to-end cooperative autonomous driving framework is introduced (UniV2X, AAAI’25).

Biomedical Discovery (BD)
  • Biomedical Agent System

    The microscopic biological system is intriguing but challenging. An all-atom framework is explored to enable consistent representation and interaction modeling across different biomolecules (PharMolixFM, TechReport’25).

  • Human-Agent Interaction System

    Recent advances in LLMs have illuminated the path toward developing knowledgeable and versatile AI research assistants across various scientific domains. Multimodal large language models are particularly promising, as they bridge the semantic gap between natural language and other modalities such as molecules, proteins, and visual information. In this context, a multimodal large language model is proposed to assist biomedical research (BioMedGPT, J-BHI’24). Additionally, the task of optical chemical structure understanding is introduced and explored to facilitate molecule-centric scientific discovery (OCSU, TechReport’25).

  • Multi-Agent Cooperation System

    Multi-agent cooperation holds great potential for solving complex scientific research tasks in an autonomous manner. To facilitate further exploration in this area, an agent platform specifically designed for biomedicine and life science is presented and open-sourced (OpenBioMed).